Red Hat AI Inference Server 3.2.5 (ROCm) is affected by multiple security vulnerabilities identified under CVE-2025-9230 and 10 additional CVEs. These vulnerabilities correspond to several CWE categories including buffer overflow (CWE-787), improper input validation (CWE-606), code injection (CWE-94), and others. The Red Hat security advisory RHSA-2025:23449 acknowledges these issues but does not list any available fixes or patches at this time. The product is not a cloud service, so remediation depends on vendor patch releases. No known exploits in the wild have been reported as of the advisory date.

The vulnerabilities collectively pose a high risk to the affected Red Hat AI Inference Server 3.2.5 (ROCm) installations. Potential impacts include unauthorized code execution, denial of service, and other security breaches due to the nature of the weaknesses (buffer overflows, improper validation, code injection). However, no active exploitation has been reported yet. The lack of an available patch means affected systems remain vulnerable until a fix is released.

As per the Red Hat advisory RHSA-2025:23449, no fixes or patches are currently available for these vulnerabilities in Red Hat AI Inference Server 3.2.5 (ROCm). Users should monitor Red Hat's official security updates for the release of patches. Until a fix is provided, consider limiting exposure of the affected server and applying any relevant workarounds recommended by Red Hat if available. Avoid speculative or generic mitigations not directly related to the advisory.